Merge branch 'main' into moat-port

Author: ZedongPeng

CMakeLists.txt

@@ -10,7 +10,25 @@ option(USE_HIP "Build with HIP for AMD GPUs" OFF)
 if(USE_HIP)
   project(cupdlpx LANGUAGES C CXX HIP)
 else()
-  project(cupdlpx LANGUAGES C CXX CUDA)
+  project(cupdlpx LANGUAGES C CXX)
+
+  # Default CUDA architectures: SASS for every current arch, PTX only for the newest one as a forward-compat fallback.
+  find_package(CUDAToolkit REQUIRED)
+  if(NOT DEFINED CMAKE_CUDA_ARCHITECTURES AND NOT DEFINED ENV{CUDAARCHS})
+    if(CUDAToolkit_VERSION VERSION_GREATER_EQUAL 13)
+      set(CMAKE_CUDA_ARCHITECTURES 75-real 80-real 86-real 89-real 90-real
+          100-real 120-real 120-virtual)
+    elseif(CUDAToolkit_VERSION VERSION_GREATER_EQUAL 12.8)
+      set(CMAKE_CUDA_ARCHITECTURES 60-real 70-real 75-real 80-real 86-real
+          89-real 90-real 100-real 120-real 120-virtual)
+    else()
+      set(CMAKE_CUDA_ARCHITECTURES 60-real 70-real 75-real 80-real 86-real
+          89-real 90-real 90-virtual)
+    endif()
+  endif()
+
+  enable_language(CUDA)
+  message(STATUS "CUDA architectures: ${CMAKE_CUDA_ARCHITECTURES}")
 endif()
 
 set(CUPDLPX_VERSION_MAJOR 0)
@@ -45,10 +63,6 @@ if(USE_HIP)
   if(NOT DEFINED CMAKE_HIP_ARCHITECTURES OR CMAKE_HIP_ARCHITECTURES STREQUAL "")
     set(CMAKE_HIP_ARCHITECTURES "gfx90a")
   endif()
-else()
-  if(NOT DEFINED CMAKE_CUDA_ARCHITECTURES)
-    set(CMAKE_CUDA_ARCHITECTURES 60 70 75 80 86 89 90)
-  endif()
 endif()
 
 # -----------------------------------------------------------------------------
@@ -112,8 +126,6 @@ if(USE_HIP)
   find_package(hipsparse REQUIRED)
   find_package(hipcub REQUIRED)
   find_package(rocprim REQUIRED)
-else()
-  find_package(CUDAToolkit REQUIRED)
 endif()
 include(FetchContent)
 

README.md

@@ -30,8 +30,8 @@ Our work is presented in two papers:
 * **Build Tools:** CMake (≥ 3.20), GCC, and NVCC (CUDA) or hipcc (ROCm).
 
 > **SpMV backend** is selected automatically at compile time based on cuSPARSE version:
-> - `cusparseSpMV` — CUDA 12.4 – 13.1 (cuSPARSE < 12.7.3)
-> - `cusparseSpMVOp` — CUDA 13.1 Update 1+ (cuSPARSE ≥ 12.7.3)
+> - `cusparseSpMV` — CUDA 12.4 – 13.2 (cuSPARSE < 12.8.1)
+> - `cusparseSpMVOp` — CUDA 13.3+ (cuSPARSE ≥ 12.8.1)
 >
 > On AMD GPUs the solver uses the `hipsparseSpMV` backend via hipSPARSE.
 

internal/cusparse_compat.h

@@ -15,21 +15,11 @@ typedef void *cusparseSpMVOpPlan_t;
 #include <cusparse.h>
 
 // cusparseSpMVOp_bufferSize was introduced in cuSPARSE 12.7.3 (CUDA 13.1 Update 1).
+// cuSPARSE 12.8.1 (CUDA 13.3.0) introduced the SpMVOp ALG1/ALG2 algorithms and
+// added a cusparseSpMVOpAlg_t parameter to cusparseSpMVOp_bufferSize/createDescr.
 // CUSPARSE_VERSION encoding: major*1000 + minor*100 + patch.
-#if defined(CUSPARSE_VERSION) && CUSPARSE_VERSION >= 12703
+#if defined(CUSPARSE_VERSION) && CUSPARSE_VERSION >= 12801
 #define CUPDLPX_HAS_SPMVOP 1
 #else
 #define CUPDLPX_HAS_SPMVOP 0
 #endif
-
-#if !CUPDLPX_HAS_SPMVOP
-// The SpMVOp types were added to cusparse.h before the functions
-// (e.g. CUDA 13.1 base has the types but not the functions).
-// Only provide fallback typedefs for cuSPARSE versions that lack them entirely.
-#if !defined(CUSPARSE_VERSION) || CUSPARSE_VERSION < 12700
-typedef void *cusparseSpMVOpDescr_t;
-typedef void *cusparseSpMVOpPlan_t;
-#endif
-#endif
-
-#endif // USE_HIP

src/mps_parser.c

@@ -408,6 +408,7 @@ lp_problem_t *read_mps_file(const char *filename)
     if (!reader)
     {
         fprintf(stderr, "ERROR: Could not open file %s\n", filename);
+        return NULL;
     }
 
     namemap_init(&state.row_map, 1024);

src/spmv_backend.cu

@@ -47,8 +47,15 @@ void cupdlpx_spmv_buffer_size(cusparseHandle_t sparse_handle,
                               size_t *buffer_size)
 {
 #if CUPDLPX_HAS_SPMVOP
-    CUSPARSE_CHECK(cusparseSpMVOp_bufferSize(
-        sparse_handle, CUSPARSE_OPERATION_NON_TRANSPOSE, mat, vec_x, vec_y, vec_y, CUDA_R_64F, buffer_size));
+    CUSPARSE_CHECK(cusparseSpMVOp_bufferSize(sparse_handle,
+                                             CUSPARSE_OPERATION_NON_TRANSPOSE,
+                                             mat,
+                                             vec_x,
+                                             vec_y,
+                                             vec_y,
+                                             CUDA_R_64F,
+                                             CUSPARSE_SPMVOP_ALG2,
+                                             buffer_size));
 #else
     CUSPARSE_CHECK(cusparseSpMV_bufferSize(sparse_handle,
                                            CUSPARSE_OPERATION_NON_TRANSPOSE,
@@ -74,8 +81,16 @@ void cupdlpx_spmv_prepare(cusparseHandle_t sparse_handle,
 #if CUPDLPX_HAS_SPMVOP
     cusparseSpMVOpDescr_t local_descr = NULL;
     cusparseSpMVOpPlan_t local_plan = NULL;
-    CUSPARSE_CHECK(cusparseSpMVOp_createDescr(
-        sparse_handle, &local_descr, CUSPARSE_OPERATION_NON_TRANSPOSE, mat, vec_x, vec_y, vec_y, CUDA_R_64F, buffer));
+    CUSPARSE_CHECK(cusparseSpMVOp_createDescr(sparse_handle,
+                                              &local_descr,
+                                              CUSPARSE_OPERATION_NON_TRANSPOSE,
+                                              mat,
+                                              vec_x,
+                                              vec_y,
+                                              vec_y,
+                                              CUDA_R_64F,
+                                              CUSPARSE_SPMVOP_ALG2,
+                                              buffer));
     CUSPARSE_CHECK(cusparseSpMVOp_createPlan(sparse_handle, local_descr, &local_plan, NULL, 0));
     *descr = (void *)local_descr;
     *plan = (void *)local_plan;